Data detection method for uplink massive MIMO systems based on the long recurrence enlarged conjugate gradient

Jawarneh, Ahlam, Albataineh, Zaid et Kadoch, Michel. 2022. « Data detection method for uplink massive MIMO systems based on the long recurrence enlarged conjugate gradient ». International Journal of Electrical and Computer Engineering, vol. 12, nº 4. pp. 3911-3921.

Prévisualisation

PDF Kadoch-M-2022-24437.pdf - Version publiée Licence d'utilisation : Creative Commons CC BY-SA. Télécharger (737kB) | Prévisualisation

Résumé

Although the mean square error (MMSE) approach is recognized to be near optimal for uplinking large-scale multiple-input-multiple-output (MIMO) systems, there are certain difficulties in the procedure related to matrix inversion. The long recurrence enlarged conjugate gradient (LRE-CG) approach is proposed in this study as a way to iteratively realize the MMMS algorithm while avoiding the complications of matrix inversion. In addition, a diagonal-approximate starting solution to the LRE-CG approach was used to speed up the conversion rate and reduce the complications required. It has been discovered that the LRE-CG-based approach has the ability to significantly reduce computational complexity. By comparing simulation results, it is clear that this new methodology surpasses well-established ways like the Neumann series approximation-based method and the Gauss-Siedel iterative method. With a small number of iterations, the suggested approach achieves near-optimal performance of a standard MMSE algorithm.

Type de document:	Article publié dans une revue, révisé par les pairs
Professeur:	Professeur Kadoch, Michel
Affiliation:	Génie électrique
Date de dépôt:	02 juin 2022 19:08
Dernière modification:	23 juin 2022 15:14
URI:	https://espace2.etsmtl.ca/id/eprint/24437

Actions (Authentification requise)

Dernière vérification avant le dépôt